An Eocene fossil scarab beetle (Coleoptera: Scarabaeoidea) from Tanzania

The Orapa fossil belongs to an extinct genus and species ( Ceafornotensis archratiras ) of the scarabaeid subfamily Melolonthinae 10 , and the three Kenyan fossils are members of three extant genera of dung beetles (Scarabaeinae


Introduction
Unusually rich and well-preserved fossil assemblages are generally known under the term 'Lagerstätte' because the information they provide is not only concerned with details of the extinct animals and plants that once lived there, thus an indication of differing levels of diversity in the past, but because it is characteristic of them that more complete records of communities are preserved. Lagerstätte additionally afford insights into palaeoecological and evolutionary relationships in the geological context.
Classic Eocene deposits in Germany, like the famous Eckfelder Maar, are considered to be Lagerstätte, but these types of fossiliferous deposits have been largely limited to the northern hemisphere. The Orapa kimberlite pipe in north-central Botswana is one of the few maar deposits in the southern hemisphere (located in eastern Botswana), and is famous for its exceptionally well-preserved compression fossils of insects and other organisms in finegrained mudstone. 1,2 Numerous maar deposits are known from central northern Tanzania (Figure 1), and at least one of them, the Mahenge maar deposit, has yielded various fossils. [3][4][5] Beetles in general, and Scarabaeoidea in particular, usually yield only a few useful characters when fossilised. Scarabaeoids are mostly compact and evenly and strongly sclerotised and when compressed and fossilised, usually only the body outline and a few legs are recognisable. In less compact beetles often only elytra become fossilised. As a consequence, well-preserved beetle fossils in sediments are relatively rare.
Fossil beetles can only be reliably identified as belonging to Scarabaeoidea (other than Lucanidae and Passalidae) by at least one of four character suites: antennae with lamellate club; pronotum and fore legs adapted for burrowing (procoxae enlarged, pronotum enlarged to include enlarged coxal musculature, protibiae dilated apically, usually with teeth on their outer margin), albeit that similar adaptations also occur in some other burrowing beetles; reduced wing venation and intrinsic spring mechanism for folding wings; prosternal intercoxal process widened apically behind procoxae. 6 The fossil record of Scarabaeoidea and a comprehensive molecular study of the origin of major beetle groups hypothesised a Triassic origin for the group 6,7 , postulated the establishment of most extant families during the Jurassic and suggested extensive radiation of groups within families during the Cretaceous, especially of phytophagous taxa in co-radiation with the expanding angiosperm flora. Yet, despite the presence of well-placed Scarabaeoidea fossils from as far back as the Middle Jurassic 8 , a total of only about 230 species have been described 6,9 ; this from one of the largest beetle groups with about 30 000 extant species and probably many thousand more extinct. Furthermore, besides a detailed study of a late Cretaceous (~ 91 mya) Melolonthinae (Scarabaeidae) species 10 , only a few Scarabaeidae of Palaeocene 9 , seven of Eocene 9,11 , another few of Oligocene age 9 , and two from the Pliocene 12 are known. Most of the fossil species of all scarabaeoid groups are from the Miocene and younger. 6,9 The Miocene was also the period when grasslands and herbivorous mammals radiated in Africa, which is when and where dung beetles (Scarabaeinae) co-radiated. 13 Three Miocene (13-12 mya) fossil species purportedly belonging to extant dung beetle genera have been described from Kenya. 14 In spite of several well-studied and extensive Eocene fossil sites (in the northern hemisphere), only a few have produced scarabaeoid fossils, and then in very low numbers: three Aphodiinae (one each from Menat, France; Baltic amber; Bognor Regis, UK) and three Melolonthinae (two species from Geiseltal, Germany and one from the Green River Formation in USA). The one described in this paper is the only Eocene scarabaeoid known from Africa.
Scarabaeoid fossils are rare in African fossil records and have only been recorded from the very rich Cretaceous deposit at Orapa in Botswana (one species: Ceafornotensis archratiras Wooley, 2016) 10 , a Miocene deposit in Kenya (three species) 14 , and a Pliocene site in northern Tanzania (two species) 12 . The specimen described in this paper lies chronologically about midway between the Orapa and Kenyan fossils, the Eocene. Earlier deposits such as the Triassic Molteno Formation yielded many insect fossils, including beetle fossils, but no scarabaeoids. 15

Geology and palaeoenvironmental reconstruction
Geologists prospecting for diamonds in the area of Singida in the early 1930s recorded the occurrence of superbly preserved fossils contained in lake beds overlying the diamondiferous kimberlite pipes. 2 The fossils were found in sediments largely consisting of shales and mudstones, originally deposited in the former crater lake. One of many in the area, the Mahenge maar deposit has yielded various fossils. [3][4][5] The Mahenge deposit has been interpreted as a small, previously roughly circular lake, about 400 m in diameter, which was formed in a kimberlite intrusion. 17,18 Generally, the formation of a kimberlite pipe begins after its eruption, when the overlying basement rock material is scattered around and a crater produced, which is subsequently surrounded by a tuffaceous cone of largely fine-grained ashes. In Mahenge, this steep-sided cone created a rather small shoreline for the original lake, made of the sediments from the surrounding pyroclastic kimberlitic rocks. 19 Thus, the centre of the Mahenge palaeolake is now characterised by well-stratified, microlaminated shales and mudstones in which the majority of the fossils are found ( Figure 2). 3,18,19 The age of the fossiliferous beds at Mahenge has been determined using radiometric methods. Originally, based on a comparison of elements in the fossil ichthyofauna, a Miocene age was suggested 20 , but this changed later to Oligocene 21 . Kimberlites from elsewhere in the Singida region ( Figure 1) have been dated using U-Pb and fission track radiometric methods, giving an age between 51 and 54 Ma for some kimberlites (Paleocene to Lower Eocene). A single zircon crystal, collected in 1996 from the stream bed at Mahenge and linked by its origin to the Mahenge diatreme, was dated at 45.83±0.17 Ma, based on 206Pb/238U radiometric measurements (Middle Eocene = Lutetian). 3 Consequently, it is assumed to be only slightly older than the palaeolake at Mahenge, because it is well known from comparable maar lakes in Europe and Africa that lacustrine sediments accumulate soon after formation of the original lake. 22 The age of the Mahenge Lagerstätte can therefore be assumed to be about 45.6 Ma old. Estimates of the sedimentation rate at Mahenge indicate that the fossiliferous deposits in the crater represent an interval ranging from 8000 to 22 700 years. 3 From palaeobotanical evidence, it has been suggested that the palaeoflora of Mahenge structurally resembled 'miombo' savanna woodlands still widespread in much of southern and East Africa, including Tanzania, and that the region was subjected to an overall dry climate with pronounced seasonality. 5,23 All the fossils are embedded as compressions in fine-laminated, pale beige, clayey-calcareous lake sediments. The rocky material seems to be extremely hard. The surfaces of the fossils are characterised at least partly by a darker film of possibly organic (coaly) or iron-containing material, varying in colour from reddish, brownish to blackish. Fine details to a size of less than 1 mm can often clearly be seen under a microscope, but processes of increasing weathering are also indicated by subsequent exuding of hardened small bubbles of iron or manganese along the body surfaces. In total, 1900 fossils have been recovered; fish make up the largest part (51%), followed by plant remains (36%) while only nine insect fossils have been identified, most of which are poorly preserved. 5 Figure 1: Overview of the Mahenge area in northern Tanzania. Potentially similar kimberlite deposits are indicated as red circles (modified after Kaiser et al. 4 ).

Examined material
The fossil specimen can be examined and interpreted from both plate and counter plate. On both pale-coloured plate sides it is shown as a dorso-ventral compression. The specimen was examined dry and wet (submerged in water) using a LEICA MZ12.5 dissecting microscope; measurements were made using digital electronic callipers and are here given in millimetres. The plates were photographed with a Canon EOS 5D camera equipped with a 100 mm macro lens. The specimen was illustrated from the photographs with the aid of vector graphics software. 24 The adult specimen is nearly complete, but parts of the head and legs are poorly preserved (Figures 2-4). The body has a cylindrical form. The fore legs are well-visible, distally broadened, and thus apparently suitable for digging. One partial outstretched metathoracic wing is visible (Figures 2, 3).

Etymology
The generic name Mahengea is based on the name of the palaeolake, Mahenge, in Tanzania from which the fossil originates.

Diagnosis
Same as the genus.

Etymology
This species is named in honour of Ian Mckay (1963Mckay ( -2022 who died suddenly and unexpectedly recently -he was an insect fossil curator, palaeoentomologist and geoscience outreach educationist extraordinaire at the University of the Witwatersrand, Johannesburg, South Africa.

Age and significance of the Mahenge fossil
Based on the geological and palaeontological evidence now known from the Mahenge site, it can be concluded that the maar and related deposits in this area represent one of the very few sampled temporal and geographical windows of freshwater and terrestrial ecosystems in sub-Saharan Africa.
Although an Eocene age is not of particular significance because of the undisputed older origin of scarabaeoids, its significance lies in the fact that few scarab fossils of older than Oligocene age have been described, and only one (Cretaceous) from Africa. It is also the only Eocene fossil scarab from Africa and one of only a few worldwide. Furthermore, its description brings to seven the total number of scarabaeoid fossils from the continent.

Possible family placement
All currently recognised families of the Scarabaeoidea (and their subfamilies) are undoubtedly older than the Eocene and the distinguishing characters outlined above could equally apply to most of these taxa. The environment was postulated to be 'savanna', so conditions conducive to habitation by almost all of the known higher taxa would have been possible and none can be excluded on ecological grounds. Consequently, we are left with little option other than to conclude that, although the fossil undoubtedly is that of a scarabaeoid species, the family should be considered incertae sedis.